While adolescent smoking rates have declined since the late 1990's, cigarette smoking among young adults (18-22) is an emerging problem. The prevalence of cigarette smoking is one of the highest in this age group and only 22% of young adults who have ever smoked regularly have quit. Despite the public health significance of smoking among young adults, our understanding of the determinants of smoking patterns and transitions in this group is very limited. In order to develop well-informed smoking prevention and treatment programs for young adults, a much more comprehensive understanding of the natural history and determinants of smoking in this population are needed. We propose to address this gap by following our established cohort of about 1,000 adolescents (initiated at age 14 years) into young adulthood (to age 22). Comprehensive data on the socio-environmental and psychological influences on smoking behavior during the adolescent phase (age 14 to age 18) will be supplemented with data collected as these individuals make the transition to young adulthood (age 18 to age 22). Specific phenotypes of interest include longitudinal patterns (trajectories) of smoking, rates of progression to regular smoking and nicotine dependence, and smoking persistence. The primary aims of the proposed study are to empirically identify the longitudinal patterns of smoking behavior (smoking trajectories) from adolescence (age 14) to young adulthood (age 22), to characterize the smoking trajectories by socio-environmental and psychological determinants, and to test mechanisms by which these determinants impact on smoking behavior. This will be one of the first studies to capture smoking transitions from adolescence to young adulthood and to evaluate smoking and its determinants in a longitudinal fashion. Results from this study can be used to inform the timing, level, type, and specific target of smoking interventions.